arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2303.10883cs.CV
收藏

低层结构分割的显式可视化算法

Explicit Visual Prompting for Low-Level Structure Segmentations

University of Macau(澳门大学);Tencent AI Lab(腾讯人工智能实验室)

Weihuang Liu, Xi Shen, Chi-Man Pun, Xiaodong Cun

摘要:

我们认为,在图像中,其中包括分割的操作部分,识别出的焦点像素,分离阴影区域,并检测隐藏的对象检测低层次结构的一般问题。虽然每个这样的主题通常都有一个特定领域的解决方案,我们表明,一个统一的方法在所有这些都表现良好。我们从自然语言处理中广泛使用的预训练和提示微调协议中得到启发,提出了一种新的视觉提示模型,命名为显式视觉提示(EVP)。与先前的视觉提示(通常是嵌入式级别的隐式嵌入)不同,我们的关键见解是强制可调参数关注来自每个单独图像的显式视觉内容,即,冻结补丁嵌入和输入的高频分量的特征。建议EVP显着优于其他参数有效的调整协议下相同数量的可调参数(5.7%的额外可训练参数的每个任务)。与特定于任务的解决方案相比,EVP还在各种低级结构分割任务上实现了最先进的性能。我们的代码可在https://github.com/NiFangBaAGe/Explicit-Visual-Prompt上获得。

英文摘要:

We consider the generic problem of detecting low-level structures in images, which includes segmenting the manipulated parts, identifying out-of-focus pixels, separating shadow regions, and detecting concealed objects. Whereas each such topic has been typically addressed with a domain-specific solution, we show that a unified approach performs well across all of them. We take inspiration from the widely-used pre-training and then prompt tuning protocols in NLP and propose a new visual prompting model, named Explicit Visual Prompting (EVP). Different from the previous visual prompting which is typically a dataset-level implicit embedding, our key insight is to enforce the tunable parameters focusing on the explicit visual content from each individual image, i.e., the features from frozen patch embeddings and the input's high-frequency components. The proposed EVP significantly outperforms other parameter-efficient tuning protocols under the same amount of tunable parameters (5.7% extra trainable parameters of each task). EVP also achieves state-of-the-art performances on diverse low-level structure segmentation tasks compared to task-specific solutions. Our code is available at: https://github.com/NiFangBaAGe/Explicit-Visual-Prompt.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑